| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
Correspondence: Address correspondence to Peter H. Ditto, PhD, Department of Psychology and Social Behavior, University of California, Irvine, 3340 Social Ecology II, Irvine, CA 92697-7085. E-mail: phditto{at}uci.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Advance medical directives • Surrogate decision makers • Medical treatment preferences
Theoretically, documenting preferences for specific medical treatment is an ideal solution to achieving patient autonomy until the end of life. If patients can convey specific wishes prior to incapacitating illness and if decision makers are able to follow these wishes exactly, patients could maintain the ability to influence their medical treatment even after they had lost the ability to speak for themselves. There are a number of problems, however, with documenting specific medical treatment preferences in advance directives. For example, research suggests that patients may have difficulty predicting their future treatment preferences (Ditto et al., 2003). Even if patients could accurately predict their preferences, a second problem is that no document could provide guidance for the infinite number of future medical possibilities (Reilly, Teasdale, & McCullough, 1994). Additionally, surrogate decision makers may be unable to convert patients' documented treatment preferences into specific treatment decisions (Ditto et al., 2001). Finally, although advance directives have become increasingly detailed in the hope that they may provide patients with greater autonomy, there is accumulating evidence to suggest that patients may, in fact, have little desire to exert this type of specific control over end-of-life medical decisions (Puchalski et al., 2000; Sehgal et al., 1992). For instance, one study found that elderly participants were more concerned with discussing the outcomes of serious illnesses—specifically how various illnesses and treatments might affect their valued life activities—than with specifying interventions to be used (Rosenfeld, Wenger, & Kagawa-Singer, 2000). Other studies have found that patients suffering from serious or terminal illness often prefer that a surrogate or physician's decisions rather than their own (previously documented) preferences be followed in the event of a disagreement between the two (Puchalski et al.; Terry et al., 1999).
This body of recent research suggests an important gap between the autonomy-centered conceptual framework that has guided advance care planning standards and the actual desires of many patients. In recognition of this gap, a growing number of researchers have begun calling for alternative approaches to advance care planning, ones that are more based on the psychological, emotional, and cognitive needs of patients (e.g., Martin, Emanuel, & Singer, 2000; Teno, 1998). For example, in recent years more attention has been paid to values-oriented advance directives (e.g., Doukas & McCullough, 1991) and to the importance of considering social and family ties in planning for end-of-life care (e.g., Allen & Shuster, 2002). Through studies such as these, it has become clear that, although some individuals may wish to document preferences for life-sustaining treatments in specific medical conditions, many individuals may prefer to delegate most or all of the decision-making control to loved ones or medical professionals. Stated another way, most people likely wish to maintain some control over certain aspects of their end-of-life medical care, but may have little interest in "micromanaging" their own death. In the present study, we sought to investigate these topics further. Specifically, our primary goal in the present study was to redirect attention from patient preferences for specific life-sustaining treatments, which have most often been the focus of past research, and instead begin to examine individuals' metapreferences, or what we refer to as process preferences. By process preferences we mean individuals' preferences about the process by which they want their end-of-life decisions made. Process preferences include individuals' wishes about whether and in what way they are interested in advance care planning, the amount of leeway they want surrogate decision makers to have in end-of-life decisions, and the people they want involved in the decision-making process. Therefore, our first goal was to simply offer some descriptive information regarding the prevalence of certain process preferences in a sample of older adults. Similarly, we sought to gain a better understanding of the values and goals people consider most important in end-of-life medical decision making.
Because surrogate decision makers play an integral role in maintaining patient autonomy at the end of life, a secondary goal of our study was to assess surrogates' understanding of patients' process preferences, values, and goals. Research shows that patient–surrogate concordance with regard to specific decisions about treatment is typically poor (Ditto et al., 2001; Layde et al., 1995). Accordingly, we sought to examine whether similar misunderstandings occur at the level of more general judgments about process, values, and goals.
Improving patients' and (especially) surrogates' comfort with end-of-life decision making is generally cited as the second key goal of advance care planning (after improving surrogates' understanding of patients' specific life-sustaining treatment wishes; see Ditto et al., 2001; Tilden, Tolle, Nelson, & Fields, 2001). Consequently, a final goal of our study was to assess patients' and surrogates' feelings of confidence in and comfort with the process of end-of-life decision making and to compare the content of their thoughts and feelings.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Procedure
Interviewers conducted interviews in the home of the patient or surrogate. Participants completed questionnaire packets by themselves, and each individual participated in a private, face-to-face interview. Interviewers presented the following questions to participants during the face-to-face interview, except where noted.
Process Preferences for Decision Making
Interviewers asked patients and surrogates a series of closed-ended questions aimed at assessing the patients' preferences regarding the following: (a) conveying treatment preferences, (b) updating preferences over time, (c) selecting decision makers, and (d) allowing leeway in decision making.
Values and Goals in Decision Making
In a questionnaire, patients were asked about the extent to which they regarded five different values (religious beliefs, dignity, physical comfort, dependency on others, and finances) as being important factors in their medical treatment decisions. We generated this list of values from findings of previous studies on individuals' considerations and preferences regarding end-of-life care (Ditto, Druley, Moore, Danks, & Smucker, 1996; Pearlman et al., 1993). Interviewers asked patients whether they agreed with, disagreed with, or were unsure about each statement expressing the importance of these five values in decision making. They asked surrogates whether they believed each of the values was important to the patient.
Interviewers asked patients to choose 3 (out of 12 possible) goals for care that they considered to be important in decision making. They then asked the patients to rank the goals as most, second most, and third most important. Surrogates performed the same task but specified the top three goals they believed patients would consider to be important in decision making. We generated the list of 12 goals from unpublished findings of an earlier phase of the ADVANCE study. During this phase, interviewers asked patients in an open-ended format to cite goals for care that they believed to be important.
Comfort With Decision Making Plans
Patients and surrogates indicated their level of comfort with current plans for decision making in a series of closed-ended questions using 5-point Likert-type scales. Interviewers asked patients about (a) comfort with their current plans, (b) confidence in the surrogate's ability to make decisions for them, (c) confidence in the surrogate's understanding of their wishes, and (d) comfort talking with family and physicians about their wishes. Interviewers asked surrogates a matching series of questions framed to assess the surrogates' own perspectives on the same topics.
Statistical Analysis
We assessed agreement between patient and surrogate process preferences, values and goals, and comfort with decision making by using two types of analyses: the kappa statistic (
) and repeated-measures t tests. In order to assess general patient–surrogate agreement, we calculated . For questions that offered (more than two) ordinal response options, we averaged patient and surrogate responses within each group and compared group means by using repeated-measures t tests. We used one-way analyses of variance (ANOVAs) and chi-square tests to compare responses by surrogate categories. We set significance at p <.05.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
=.13, p <.01) and there were no significant differences between group means (p >.05).
|
=.04, p >.05).
Over half of the patients (57%) and surrogates (61%) believed that the best approach to documenting preferences was one that included written requests and verbal communication with family members. Of those who did not choose both methods, more believed that patients should talk to decision makers about preferences than write treatment preferences in an advance directive. Slightly more than half of the pairs agreed on which approach was best for conveying wishes ( =.14, p <.01); however, most disagreements (63%) occurred when either the patient or surrogate thought the patient should both write and talk about preferences whereas the other thought the patient should only talk about the preferences.
Of the patients and surrogates who desired an advance directive, the largest proportion (50% and 44%, respectively) preferred one that contained only general statements about values or goals for care that patients would want guiding medical decisions. Fewer patients and surrogates desired an advance directive that included both general statements and precise treatment directions, and fewer still preferred one that included only precise directions regarding specific medical treatments. Patient–surrogate agreement was relatively low (45%) but better than expected by chance ( =.13, p <.01). The most common type of patient–surrogate disagreement (52%) occurred when either the surrogate or patient desired an advance directive that included both value statements and precise directions whereas the other desired one that included only value statements.
The majority of participants believed that surrogates should have some leeway to override the patients' preferences if they believed it would be best for patients. A substantial proportion of each group believed that surrogates should have "a lot" or "complete" leeway, whereas only 9% believed the surrogate should have "no" leeway. Patient–surrogate agreement was somewhat low (40%) but better than chance alone ( =.12, p <.001). When we dichotomized responses into two groups—those wanting surrogates to have leeway and those not wanting leeway—agreement increased to 84%. On average, patients (M = 2.77, SD = 1.15) believed surrogates should have significantly more leeway in decision making than surrogates (M = 2.40, SD = 1.05) believed they should have, t(320) = 6.08, p <.001.
When interviewers asked patients how they wanted decisions made for them, most expressed a preference for having several people talk and then one person (presumably the surrogate decision maker) make the final decisions alone. Fewer patients and surrogates believed that one person should decide completely alone or that several people should make the decisions together. Slightly fewer than half of the pairs preferred the same decision-making strategy; however, this agreement was better than chance ( =.14, p <.001). The most common type of patient–surrogate disagreement (30%) occurred when the patient preferred that only one person decide (without talking to others) and the surrogate preferred that several talk and one decide. Overall, patients were slightly more likely than surrogates to prefer that one person decide alone (25% vs 15%, respectively).
When specifying which individuals should be included in decision making, a majority of patients and surrogates selected family members and physicians. A few wanted clergy involved and a few wanted "others" involved. Surrogates wanted significantly more people involved in decision making (M = 2.40, SD =.65) than did patients (M = 2.25, SD =.63), t(202) = –2.63, p <.01.
Values and Goals in Decision Making
Table 3 presents patients' and surrogates' responses to the questions about important values in end-of-life decisions. With one exception, a majority of patients and surrogates believed that each of the values presented were important. Although most patients valued "not being a financial burden," few surrogates believed it was important. "Having a dignified death" was the most highly endorsed value by both patients and surrogates.
|
The goals for care that were selected as being most important can be seen in Table 4. Retaining the ability to think clearly and remember important things was cited as the most important goal by the largest proportion of patients (28%) and surrogates (25%). The second most-endorsed goal was retaining the ability to take care of one's own personal needs.
|
=.09, p <.001). Given three tries, fewer than half of the surrogates (48%) were able to correctly identify the patient's most important goal for care.
Comfort With Decision-Making Plans
Patient and surrogate comfort with plans can be seen in Table 5. The vast majority of patients and surrogates felt "pretty" or "extremely" comfortable with the patients' current plans for decision making. Only 9% of the patients and 15% of the surrogates felt less than pretty comfortable. Patient–surrogate agreement was better than chance ( =.14, p <.05). Approximately one third (32%) of the surrogates felt less comfortable than their patient partners, and a mean comparison revealed that surrogates were significantly less comfortable with patients' plans (M = 4.16, SD =.86) than were patients themselves (M = 4.32, SD =.74), t(320) = 2.98, p <.001.
|
=.14, p <.001 for "talking with family" and =.10, p <.05 for "talking with physician"), mean comparisons suggested that surrogates viewed the patients as more uncomfortable talking with family members (M = 1.47, SD =.76) than the patients reported being (M = 1.36, SD =.68), t(320) = 2.98, p <.001.
A majority of patients and surrogates reported having confidence in the surrogate's decision-making abilities; however, mean comparisons showed that patients were more confident in the surrogates' decision-making abilities (M = 4.49, SD =.69) than were surrogates themselves (M = 4.24, SD =.67), t(321) = 5.21, p <.001. A similar pattern emerged when patients and surrogates were asked how well they believed the surrogates understood the patients' wishes. A majority in both groups believed the surrogate understood the wishes at least "pretty" well, whereas less than 10% of each group believed the surrogate understood the patients' wishes "somewhat," "not very," or "not at all" well. Patient–surrogate agreement was significantly better than chance ( =.19, p <.001); however, surrogates believed they understood patient wishes significantly less well (M = 4.34, SD =.67) than patients believed they did (M = 4.50, SD =.70), t(321) = 3.72, p <.001.
Comparison of Spouse and Child Surrogates
To examine the possibility that the nature of the patient–surrogate relationship might affect process preferences and attitudes toward end-of-life decision making, we repeated the aforementioned analyses comparing responses by the two most common surrogate categories (i.e., spouses vs adult children). In general, we found that the two types of surrogates showed similar patterns of responses, although a number of interesting differences emerged. Spouses were significantly less likely than children to believe that recording wishes in an advance directive was necessary (Ms = 2.36 vs 2.60), F(1, 295) = 6.80, p <.05, and they tended to believe they should have slightly more leeway in decision making (M = 2.48) than children believed they should have (M = 2.25), F(1, 295) = 3.17, p =.08. More children than spouses believed that patients should update their advance directives regularly (70% vs 53%), 
2(1, N = 288) = 7.78, p <.01, and children were more likely than spouses to endorse writing and talking (combined) as the preferred method of conveying preferences (77% vs 57%), 2(3, N = 296) = 11.81, p <.01. Regarding values in decision making, spouses were more likely than children to believe that financial concerns were important (Ms = 1.92 vs 1.69), F(1, 294) = 4.34, p <.05. Additionally, patients believed that children understood their wishes better than did spouses (Ms = 4.61 vs 4.42), F(1, 306) = 4.47, p <.05, and they were marginally more confident that children's decisions would match their own (M = 4.60) than would those of their spouses (M = 4.44), F(1, 307) = 3.51, p =.06. We found no differences in patient–surrogate agreement with regard to process preferences, values, goals, or feelings of comfort with decision making (all ps >.05).
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Along with findings from previous research on patient preferences, our findings suggest an ongoing mismatch between some of the common standards in advance care planning methods and the needs and desires of many patients. For example, patients consistently appear to place a high priority on communicating their preferences verbally (Sehgal et al., 1992), whereas the law generally regards oral advance directives as less valid than written ones (Cruzan v. Harmon, 1988). Patients also seem more interested in conveying their personal values and goals for decision making than specifying treatment interventions (Rosenfeld et al., 2000; Sehgal et al.), but the formats of most advance directives do not reflect this priority. Finally, patients repeatedly show a desire to allow trusted others some level of flexibility in decision making (Puchalski et al., 2000; Sehgal et al.; Terry et al., 1999); however, this preference is normally not conveyed in detail at any stage of advance care planning. In light of legal rulings favoring specificity of documented preferences and in light of the ambiguity inherent in planning for unknown future events, it is easy to see why advocated planning methods have pushed patients to document their preferences with more and more medical detail and rigidity. However, it is becoming increasingly clear that most patients do not want to micromanage their end-of-life care to the extent that most of these methods imply. Although some medical care providers, with adequate experience or insight, may encourage individual patients to engage in a more thorough process of planning for the end of life, this process-oriented approach to planning has not received the formal attention and legal endorsement that the traditional treatment-based approach has received. If patients are only aware of rigid treatment-based documents for guiding care at the end of life, their autonomy may actually be hindered, thus defeating the intended purpose of advance directives. It is important that medical care providers, researchers, and policy makers recognize this unintended possibility.
Diversity of Individual Process Preferences
A second important point highlighted by our findings has to do with the substantial variability of preferences expressed by the patients in our study. For example, although we did find that patients largely desired to grant surrogate decision makers leeway in decision making, the amount of leeway desired varied substantially between individuals. Some patients wanted the surrogate to have complete leeway in decision making, whereas others wanted the surrogate to have little. A small group insisted that the surrogate have no leeway to override the patient's advance directive. Although it is noteworthy that most patients desired some leeway, it is important not to generalize the preferences of the majority to all patients as preferences for leeway appear to be quite diverse. Additionally, previous research suggests that patients may have individual preferences for the types of decisions that surrogates may override versus the types that should strictly follow a patient's documented wishes (Terry et al, 1999). In any case, the extent of flexibility that patients desire surrogates to have in decision making is important information that should be clearly conveyed to designated decision makers. Some researchers have even suggested that all patients with advance directives specify where they would like the "locus of decision making" to reside (Terry et al.).
In addition to beliefs about surrogate leeway, patients also displayed diverse opinions on various aspects of the planning and decision-making process. For example, some patients wanted to complete advance directives only once whereas others wanted to update them frequently. Some wanted decisions to be made by a team of people and others wanted decisions made by only one trusted other. Perhaps the most important individual preferences were the values and goals patients endorsed as being most important in guiding end-of-life decisions. The values and goals most often chosen by patients in our study were consistent with past research in that they were largely concerned with cognitive functioning, independence, and dying with dignity (Pearlman et al., 1993; Sehgal et al., 1992); however, it is important to recognize that not all patients in our sample valued the same considerations. Each of the values we presented to patients was considered important by some and unimportant by others. Likewise, each of the 12 goals for care was selected as being the most important to at least a few patients. Not even one of the values or goals was considered to be important by all of the respondents. Because each patient has a unique set of personal values and goals that he or she regards as central to life and life-sustaining treatment decisions, it is essential that all patients clearly explain their values and goals to surrogates.
Patient–Surrogate Understanding
Previous research on patient–surrogate understanding has generally shown that surrogates are not proficient in predicting patients' specific wishes for medical treatment (Ditto et al., 2001; Seckler, Meier, Mulvihill, & Cammer-Paris, 1991). Findings from the present study suggest that surrogates may generally be lacking knowledge about patients' preferences for other aspects of decision making as well. For instance, there were high rates of patient–surrogate incongruities on several of the endorsed values in decision making. As a whole, surrogates in our sample tended to be unclear about whether religious beliefs were important to their patient partners, and they largely undervalued avoiding dependence and being a financial burden on others. Children of patients were especially likely to underestimate the importance their parents placed on financial concerns.
When surrogates were asked to choose the one goal that they believed their patient partners would choose as being most important, less than one fourth of the surrogates cited the correct goal. When surrogates were given three chances to cite the most important goal, still, fewer than half of the surrogates were able to do so. Such lack of concordance between patients and surrogates presents a major challenge to the surrogates' abilities to make decisions in the way that patients desire. If patients want surrogates to base decisions on what they know about the patients' values and goals for care, it is imperative that surrogates understand these things from the patients' perspective. It would benefit patients and surrogates alike if advocated planning methods encouraged patients to actively convey guiding values and treatment goals for decision making to surrogates.
Our findings also indicate that patient–surrogate communication could be improved with regard to other aspects of the decision-making process. For example, fewer than half of the surrogates in our study knew the extent of leeway that the patients wanted them to have in decision making, and most were unsure of the patients' preferences for how decisions at that time should be made. In most cases of patient–surrogate incongruity, surrogates asserted significantly less leeway to override preferences than patients desired to grant them. Surrogates also desired to include more people in the decision-making process than the patients desired to be included. This finding is important because it suggests that patients may not be communicating to surrogates some of their most basic assumptions and preferences about the way they envision the final stages of decision making to unfold. In contrast to patient–surrogate understanding on more difficult topics such as whether or not to discontinue artificial feeding under a variety of medical prognoses, a good level of understanding about a patient's process preferences may be more achievable. Not only would this understanding allow surrogates to better fill the role in decision making that patients had intended, but it could also alleviate the stress surrogates may feel from (inaccurately) believing that patients want them to know and be able to apply an overwhelming array of treatment preferences to unanticipated circumstances. For example, patients in our study reported particular confidence in the surrogate's understanding of their wishes when the surrogate was a child rather than a spouse. Ironically, children surrogates expressed a greater desire for specific guidance from the patient (e.g., stronger endorsement of written advance directives and weaker endorsement of surrogate leeway) than spouse surrogates. If children surrogates knew of their parent's confidence in them, some of the uncertainty and discomfort associated with end-of-life decision making might be ameliorated. Of course, it is possible that no matter how clear patients were in conveying their process preferences to surrogates, surrogates would still feel ambivalence toward accepting decision-making authority. The role a surrogate plays in a loved one's care at the end of life is not an easy one, and a surrogate's lack of confidence is understandable. In any case, our findings indicate that both patients and surrogates may benefit from increased communication about the surrogates' role in decision making.
Study Limitations
There are several limitations of our study that should be addressed. First, the vast majority of participants were European American. Because attitudes toward end-of-life medical treatments have been found to vary across ethnic groups (e.g., Hopp & Duffy, 2000), attitudes pertaining to process preferences may vary as well. In addition, the sample was a subset of a larger sample recruited into a longitudinal study, and the participants who remained in the study at the time of data collection for the present study tended to be slightly younger, healthier, and wealthier than the participants who dropped out after an earlier phase. It is possible that some of the answers and opinions expressed in the present study may have been influenced by the participants' involvement in earlier phases of the study. However, there is no compelling reason to believe that this subset of participants would be less likely to endorse the concept of micromanagement in end-of-life decision making nor that they would have more diverse opinions about the preferred process of decision making. In addition, it is important to note that there existed substantial patient–surrogate misunderstandings about the patient's general process preferences, as one might have expected increased discussion and familiarity with advance directives to result in better patient–surrogate communication or understanding.
Conclusion
There are many types of decisions that must be made at the end of life. In the existing literature on advance directives, there has been a disproportionate focus on specific medical treatment decisions, leaving other types of decisions, such as those about the preferred process for planning and decision making, largely unaddressed. Findings from our study indicate that these unaddressed process preferences are an important component of advance care planning. Future research should investigate other aspects of the decision-making process that patients may want to influence. In addition, more research is needed on patient–surrogate communication of these issues to promote broader surrogate understanding of patients' process preferences.
In practice, advance care planning standards could benefit from allowing patients more flexibility in conveying their preferences to others. Our findings suggest that many patients do not desire to micromanage the course of medical treatment they receive at the end of life and that being encouraged to convey different types of information and preferences in advance directives may be advantageous. Specifically, if advance directives were used to convey information about how much authority patients want surrogates to have in decision making, and to communicate a wider scope of values and goals the patients feel are most important in life, patients could potentially gain the level of control they desire over their final days and surrogates may be more successful, and feel more comfortable, honoring the wishes of the patients.
|
Footnotes |
|---|
1 Department of Psychology and Social Behavior, University of California, Irvine.
2 Department of Psychology, Kent State University, Ohio.
3 Family Practice Center of Akron, Summa Health System, Akron, Ohio.
Decision Editor: Linda S. Noelker, PhD
Received for publication January 23, 2004. Accepted for publication June 21, 2004.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
4206, 4751 of the Omnibus Reconciliation Act of 1990, Pub.L. No.101-508 (1990).This article has been cited by other articles:
 |
|
 |
|
  B. Bergman-Evans, L. Kuhnel, D. McNitt, and S. Myers Uncovering Beliefs and Barriers: Staff Attitudes Related to Advance Directives American Journal of Hospice and Palliative Medicine, October 1, 2008; 25(5): 347 - 353. [Abstract] [PDF]
|
|
|
|
 |
|
 E. P. McCarthy, M. J. Pencina, M. Kelly-Hayes, J. C. Evans, E. J. Oberacker, R. B. D'Agostino Sr.,, R. B. Burns, and J. M. Murabito Advance Care Planning and Health Care Preferences of Community-Dwelling Elders: The Framingham Heart Study J. Gerontol. A Biol. Sci. Med. Sci., September 1, 2008; 63(9): 951 - 959. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Workman A communication model for encouraging optimal care at the end of life for hospitalized patients QJM, December 1, 2007; 100(12): 791 - 797. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Volandes, L. S. Lehmann, E. F. Cook, S. Shaykevich, E. D. Abbo, and M. R. Gillick Using Video Images of Dementia in Advance Care Planning Arch Intern Med, April 23, 2007; 167(8): 828 - 833. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. R. Fried, A. L. Byers, W. T. Gallo, P. H. Van Ness, V. R. Towle, J. R. O'Leary, and J. A. Dubin Prospective Study of Health Status Preferences and Changes in Preferences Over Time in Older Adults. Arch Intern Med, April 24, 2006; 166(8): 890 - 895. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Zimmerman, P. D. Sloane, C. S. Williams, P. S. Reed, J. S. Preisser, J. K. Eckert, M. Boustani, and D. Dobbs Dementia Care and Quality of Life in Assisted Living and Nursing Homes Gerontologist, October 1, 2005; 45(suppl_1): 133 - 146. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Minerva BMJ, February 26, 2005; 330(7489): 488 - 488. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
